Blockchain project settles cross-border payment

Singapore’s central bank sent a payment to Canada using blockchain technology last week, in a clear signal that the technology has value – as long as you’re realistic about it.

The Monetary Authority of Singapore (MAS) sent $105 Singapore dollars to the Bank of Canada (BoC) in a proof-of-concept project that inches them closer to solving one of banking’s biggest headaches: cross-border payments and settlements.

In a November 2018 report on cross-border interbank payments and settlements, the two organizations and the Bank of England detailed the challenges of settling transactions between banks in different countries. Banks must navigate an array of hurdles including anti-money-laundering and know-your-customer regulations.

If a bank has no presence in the recipient country, it must also rely on another intermediary bank to process the payment on its behalf, in what’s known as the correspondent banking model. All the parties will have their own legacy systems that make it difficult to process the transaction uniformly. It is an expensive process that can take several days, and parties never quite know when the money will arrive.

The biggest problem is counterparty risk – when a bank sends money via an intermediary to buy something, it can‘t be certain that the intermediary will deliver the funds, or that the other bank in the transaction will hold up its end of the bargain.

Reducing counterparty risk

BoC and MAS wanted to use the blockchain to settle payments while reducing counterparty risk. Each organization already had its own distributed ledger for processing the clearing and settlement of payments and securities domestically. In 2016, BoC created Project Jasper, while MAS created Project Ubin. This latest project brought the two distributed ledger technologies together so they could collaborate on transactions.

The project still needs an intermediary that has a presence in both the sending and receiving countries, because the intermediary is the only party that carries both the sending and receiving banks’ currencies and can process payments in both countries. The difference is that no funds need change hands between this intermediary and the sending and receiving banks before the entire transaction is complete, which reduces the risk.

Instead, the whole multi-step exchange relies on ‘atomicity’, which is a condition where a group of transactions in a chain must all succeed. If one transaction fails, then they all fail, rolling back the exchange of assets to the beginning.

BoC and MAS achieved atomicity using Hashed Time Locked Contracts (HTLCs). If Alice wanted to buy something from Bob for $1,000, it would work like this:

1. Bob generates a random number and uses it to create a hash, which he sends to Alice.
2. Alice sends the $1,000, along with the hash that Bob sent her. To unlock that payment, he must show her the data he used to produce the hash.
3. Bob produces the data and unlocks the payment.

If these steps don’t happen in time, all transactions fail.

Because this is all happening on the blockchain, the banks encode this HTLC in a smart contract (a computer program that runs on all participants’ computers on the blockchain). The smart contract checks all conditions are met before unlocking the payment.

The BoC/MAS exchange goes like this:

1. The Canadian bank uses a secret (a random number) to generate a hash. It sends the hash to the Singapore bank.
2. The Singapore bank deposits the cash it wants to send to the Canadian bank in an escrow account on the blockchain. It then sends the hash to the intermediary bank’s Singapore branch.
3. After checking the smart contract to ensure that the funds are in the escrow account on the blockchain, the intermediary bank sends the hash to its Canadian branch. The intermediary bank’s Canadian branch deposits the funds in another escrow account on the blockchain. Then it sends the HTLC to the Canadian bank.
4. The Canadian bank verifies that the locked amount is correct and then uses the original secret to unlock the funds from the escrow account. It also sends the secret to the intermediary bank’s Canadian branch.
5. The intermediary branch in Canada shares the secret with its Singapore branch.
6. The intermediary’s Singapore branch then uses it to unlock the funds in the escrow account there.

A SGD$105 payment between two parties is a good start, but as the two central banks note in their report on the project, it still needs work if it is to scale. It would need something like gateway nodes that act as service nodes for their network participants, or a central connector between networks.

It’s all very exciting, but the Ripple cryptocurrency-based cross-border payment network has been doing more or less the same thing for years now. However, operating their own payment network rather than using a third-party payment service would presumably give central banks autonomy.

The whole thing also shows that the blockchain isn’t a white elephant. The technology slipped into what‘s known as the trough of disillusionment on Gartner’s hype cycle last August. This trough is part of a natural transition as people realize that a single technology won’t solve all of humanity’s problems.

From here, blockchain faces a slow, steady climb back into favour as people realize that it has some value for certain use cases, as long as you view it sensibly.